As a step to reducing potentially carcinogenic exposures for the gastrointestinal (GI) tract, novel semi-permeable magnetic microcapsules are shown capable of revealing the presence in vivo of carcinogen metabolites, nitrosating agents and potentially genotoxic cross-linking agents; trapping is shown to be highly dependent upon diet in an epidemiologically-relevant manner. These versatile microcapsules could ultimately be used in molecular and nutritional epidemiology; the present proposal is focussed on establishing and validating them both for the aforementioned agents and as-yet unknown endogenous DNA- damaging agents. In the next three years, this project continuation has three aims: (a) To validate the microcapsule system in vivo using model carcinogens, nitrosating and cross-linking agents, both regarding adduct formation and its relationship to several biological end- points within and exterior to the intestine. (b) To further develop microcapsules containing target moieties (nucleic and amino acids) that can subsequently be cleaved out with their adducts for ultra-sensitive analysis (32 P-post-labelling, mass spectrometry, etc.); this takes advantage both of microcapsules as a stable vehicle for gut transit and recent analytical advances for DNA/hemoglobin adducts. (c) To apply the microcapsule to a set of GI risk-related substances (fecapentaenes, N-nitrosated bile acids, cross-linking agents) and utilise our discovery of dietary influence on trapping of carcinogens. Monitoring of the GI cavity is of high priority due to known high risk, high exposure and seemingly facile, non-invasive investigation, but no other method is available and so this microcapsule system could have many future applications within the GI tract (and e.g. environmental applications). The methodologies proposed above seem not to include any crucial unexplored steps, and mainly consist of validation and extension of preliminary work performed under previous support.